At first I wanted to write a review of QuarkChain to get some points for their crowdsale lottery. Once I read the whitepaper though I became so thrilled about the project that writing a review is something that I actually want to do for the sake of it.
What is the trilemma problem?
There are three main issues to solve when building a blockchain - security, decentralization and scalability. It is insanely hard to solve all three in the same time, you have to make some compromises. For example Bitcoin and Ethereum are very secure(yes, even though Ethereum smart contracts has been hacked, that is because of the people who wrote them, not because of the underling network), but they lack scalability. EOS on the other hand is very scalable but lacks decentralization and some will argue - security. They are many up-and-coming blockchains that claim to have solved decenralization and scale, but they will very likely have security problems(think Credits and the likes).
QuarkChain uses sharding to solve the scalability issue. Ethereum wants to implement sharding, but it will take years before they get there. It is very difficult to re-design such a big system that has been running for years already. QuarkChain's advantage is that they designed everything from scratch with decentralization and security in mind.
So what is sharding in the blockchain realm? Its partitioning the network in multiple blockchains. It is a very difficult task even for the best engineers out there, because of its challenges:
- making cross-shard transactions - with QuarkChain users will be able to seemlessly move funds from one shard to another.
- single shard take-over. This is a very serious security issue: since there will be a huge amount of shards, every shard will have a small amount of Hash power, compared to the whole network. So how do you prevent someone from making an 51% attack in a single shard? QuarkChain uses a root chain for solving this issue. Every transaction is confirmed there to ensure the security of the whole system.
To make an attack on the system one needs to have 51% of the hash power on the root chain, or 25% of the power of the system. You may ask the question: isn't that a lot less than Ethereum/Bitcoin and thus insecure? QuarkChain has a solution that in the same time insures decentralization: incentivizing parties Not to form pools and making sure that smaller players can participate in the network as well. Since QuarkChain uses PoW algorithm, to have a full validator node in the system one may need to have a huge resource, since validating 1m TPS is very heavy task. QuarkChain solves this problem by introducing clusters. A cluster is a set of nodes(for example PCs) that will validate different subsets of chains. Every cluster will be effectively a full node and every miner will receive his reward in timely manner, thus eliminating the need of pooling. In a system with no pools it will be much harder to aquire a large percentage of the computational resources and the 51% attack is effectively harder to occur.
QuarkChain will make account management a piece of cake. One account to rule them all - you will need just one set of keys to control funds on ALL shards in the network!
QuarkChain will run EVM - the Ethereum virtuam machine. Thus any developer who has build on Ethereum can easily build on QuarkChain, and moreover any smart contract/ Dapp that is build on Ethereum can seemlessly transfer to QuarkChain.
I am not a technical person, but I can see that QuarkChain has a huge potential to solve the trilemma. Their solution is simple, build on incentives, with decentralization, security and scalability in mind. So thrilled about the future of blockchain when I read a whitepaper like this one.